Skip to main content
Forest Ecosystems

Table 2 The response and predictor variables and the principles of relating these. The modelling principles (MSN: Most Similar Neighbor, SUR: Seemingly Unrelated Regression) are detailed in Section 2.4

From: On the potential to predetermine dominant tree species based on sparse-density airborne laser scanning data for improving subsequent predictions of species-specific timber volumes

  

Modelling principle

Response variables

Predictor variables1

MSN

SUR

- Total plot volume

- Species-specific volumes

- ALS-based CBH estimate

- Maximum, mean, standard deviation and proportion

- Percentiles 5, 10, 20, …, 90, 95

- Densities 5, 10, 20, …, 90, 95

- Mean and standard deviation of intensity values2

NN search based on canonical correlation analysis between all response and predictor variables. The dominant species are included as restrictions to the NN search.

System of linear regression equations based on 1–2 ALS features and a categorical predictor indicating the dominant species on the plot.

  1. 1Computed using height values > 2 m ground threshold
  2. 2 Computed separately based on only, first-of-many and both the return types
Back to article page